PROJECT SUMMARY: Liver cancer is a common cause of cancer death worldwide, estimated to be responsible for over 700,000 deaths annually. There are multiple risk factors for liver cancer including alcohol abuse, chronic infection with hepatitis B or C viruses, cirrhosis, non-alcoholic fatty liver disease, and consumption of aflatoxin-contaminated food. Liver damage and inflammation are common elements inherent to these risk factors. Cancer genome sequencing data has established tumor type-specific mutational spectra for a variety of tumors including hepatocellular carcinoma (HCC). In some cases (lung cancer in smokers, skin cancers, certain urothelial carcinomas), the mutational patterns can be used to infer the involvement of exposures and certain types of DNA adducts in cancer etiology. For HCC, a dominant mutational signature has been found and is present across different geographic regions. This signature consists of AG transitions, which are rare in other tumor types. The AG mutations are particularly enriched on the non-transcribed strand of expressed genes and increase with gene expression levels, a phenomenon referred to as transcription-coupled DNA damage. We hypothesize that the adducts initiating the liver cancer AG signature preferentially form at single-stranded regions in transcribed genes called R-loops. We further hypothesize that lipid peroxidation product-derived bifunctional aldehydes formed in the liver as a consequence of inflammation and oxidative stress produce etheno- adenine DNA adducts selectively at these single-stranded regions and cause AG mutations as a consequence of lesion-tolerant DNA polymerase bypass which inserts cytosine across the etheno-A adducts. We propose two Specific Aims to test our hypothesis: In Aim 1, we propose to develop a novel method for genome-wide adductomics and will focus on etheno-A adducts induced by bifunctional aldehydes. We will determine if these adducts are more frequent in genomic sequences that are highly transcribed and form R-loops. In Aim 2, we will establish the mutational spectra of lipid peroxidation product-derived aldehyde adducts in immortalized hepatocyte single cell clones. The sequencing data will be used to derive a mutational spectrum of the treated cells in comparison to control solvent-treated cells. We will thus determine if the tested bifunctional aldehydes are capable of producing AG mutations, similar to those found in HCC genomes. This application is aimed at developing a test system for identifying the causative mutagens for liver cancer including their respective DNA adducts and induced mutations. Method development is a significant component of this application.